Project:
This dataverse dataset contains materials for the replication of �Campaign Donations, Judicial Recusal, and Disclosure: A Field Experiment.� 
Because the inclusion of all covariates used in the analysis of the paper would allow for the identification of subjects (judges, attorneys, and parties), the data included in this dataverse dataset only includes subject numbers, treatment assignments, treatment probabilities, and key experimental outcomes. For access to a more complete version of the data, please contact the dataverse administrator (Dane Thorley).
Contents:
1. Readme (this file)
2. Replication Code (�Krasno et al Recusal Replication Code (7.27.2020).R�: This is the full code used to conduct the analysis presented in the paper. (NOTE: This code will not work with the de-identified dataset included in this dataverse dataset).
3. De-identified Replication Data (�Krasno et al Recusal Replication Data (Experimental Cases) (7.27.2020).csv�: This is a de-identified dataset used to conduct the main analysis in the paper (Tables 1, 2, E1, E2). It only includes subject numbers, treatment assignments, treatment probabilities, and key experimental outcomes.
4. De-identified Replication Data (�Krasno et al Recusal Replication Data (NonExperimental Cases) (7.27.2020).csv�: This is a de-identified dataset used to conduct the spillover analysis in the paper (Table E3). It only includes subject numbers, spillover eligibility, spillover assignment probabilities, and key experimental outcomes.
5. Variable Dictionary (�Krasno et al Recusal Replication Variable Dictionary.csv�): This is a variable dictionary for all the data used to conduct the primary analysis in the paper. 
6. Replication Output (�Krasno et al Recusal Replication Output (7.27.2020).txt�): This is the output of the Replication Code (NOTE: This output reflects the replication code applied to the full dataset, including the variables that were removed from the De-identified Replication Data to avoid the identification of subjects).
